Monday, November 22, 2010

Afterthought on afterimages: Green flashes and green lights

Some experiments are fun and not painstaking. Here is one from my student days ...

by Michael H. Brill, Datacolor

Afterimages from bright lights are usually undesired and hardly ever helpful, but they can tell you something about the visual system. Here’s an experiment you can try at home, that shows a remarkable interaction of your two eyes during an afterimage. You will need a strong pen-light overlaid by a green filter, a strong, directed white light (such as a 500-watt fiber-optic projector), a magnifying glass lens, half a ping-pong ball (placed over the right eye), and a white wall under artificial light (incandescent will do). The problem, as you do the experiment, is to explain the afterimage effect.

First of all, hold the green pen-light at arm’s length and flash it into your left eye. Then look away at the white wall. The long-lasting dark (negative) afterimage will look magenta at first and then turn bluish purple. This happens whether or not your right eye is open. Closing the left eye or dimming the light on the white paper makes the afterimage turn bright green (positive).

Now repeat, but when you look at the wall with your left eye, shine the projector into your right eye through the half ping-pong ball, producing a uniform field that won’t distract you from the left-eye’s afterimage. I think you will see the afterimage in the left eye flash bright green for a second and then return to a dark purple appearance. By the way, the projector should be about 3 feet away from your right eye, and you should look at it only through the ping-pong ball.

As long as the negative afterimage persists, the green flash can be elicited repeatedly by turning the projector light on and off. The flash is the same color as the light occasioning the afterimage (in this case green). The green flash cannot be attributed to stray light entering the left eye from the light producing the uniform field on the right eye: More light in the left eye just makes the afterimage appear darker and more purple. On the contrary, the flash effect is similar to the polarity reversal that happens when light is dimmed in the left eye.

So why is this happening, and how can you prove it?

Give up? Well, it seems that when you turn on the light in the right eye, the right-hand iris contracts, and that causes the left-hand iris to contract as well. That dims the light from the white wall by decreasing the pupil diameter. (This effect will be strongest in young people who still have some action in their irises.)

How can I show this? Repeat the above experiment, instead of looking at the white wall, look at a distant white light source through the magnifying glass. Position the lens so its near focal point is in the plane of the pupil, sends light through the middle of the pupil without being affected by the iris. You’ll know you have the right distance when the image of the light source floods your whole retina. Now turn on the projector to the right eye, and lo! The green flash will not appear.

It might appear that you need about eight arms to do this, so it is not as casual an experiment as I have misled you to believe. But it is not quantitatively demanding, and I published it for a small audience at MIT without incident [1]. As of this year, you can read any of the progress reports on the Web. It costs no money, and the complete obscurity of this unrefereed publication is balanced by its refreshing availability to all, without passing a toll gate. Academic freedom has turned inside out, and we have found an unexpected place where the green light is flashing.

M. H. Brill, Binocular afterimage effect, MIT Research Lab. of Electronics Progress Report, PR 120, pp. 168-169 (1978). http://dspace.mit.edu/bitstream/handle/1721.1/56698/RLE_PR_120_XXVI.pdf?sequence=1.